新的改进IEEE 802.11 DCF性能的退避机制

分布式协调功能DCF是IEEE802.11标准最基本的媒体接入方法,它的核心是载波检测多址接入/冲突避免（CSMA/CA）机制,通过退避算法,减少碰撞的概率。提出了一种新的退避机制改进IEEE802.11DCF饱和吞吐量性能,建立了三维马尔可夫链网络模型详细研究分析,同时利用NS2对所提出的机制进行仿真,比较了改进后的802.11DCF饱和吞吐量与原802.11DCF的饱和吞吐量的大小,仿真结果证明了算法的准确有效。     

Optimal Constant-Window Backoff Scheme for IEEE 802.11 DCF in Single-Hop Wireless Networks Under Finite Load Conditions

Existing backoff scheme’s optimization of IEEE 802.11 DCF MAC protocol consider only saturated networks or asymptotic conditions. In real situations, traffic is bursty or streamed at low rates so that stations do not operate usually in saturated regime. In this work, we propose and analyze a backoff enhancement for IEEE 802.11 DCF that requires information only about the network size and that is quasi-optimal under all traffic loads. We first analyze the performance of DCF multiple access scheme under general load conditions in single-hop configuration and we provide an accurate delay statistics model that consider the self-loop probability in every backoff state. We prove then the short-term unfairness of the binary exponential backoff used in IEEE 802.11 by defining channel capture probability as fairness metric. Motivated by the results on fairness, we introduce the constant-window backoff scheme and we compare its performance to IEEE 802.11 DCF with Binary exponential backoff. The quasi-optimality of the proposed scheme is proved analytically and numerical results show that it increases, both the throughput and fairness, of IEEE 802.11 DCF while remaining insensitive to traffic intensity. The analysis is then extended to consider the finite queuing capacity at nodes buffers using results from the delay analysis. NS2 simulations validate the obtained results. Institut Eurecom’s research is partially supported by its industrial members: BMW Group Research & Technology—BMW Group Company, Bouygues Telecom, Cisco Systems, France Telecom , Hitachi Europe, SFR, Sharp, STMicroelectronics, Swisscom, Thales.     

A Novel Event Model for Performance Analysis of IEEE 802.11 DCF

After considering the memory effect among series events occurring on the channel, we propose a novel event model to analyze the channel status more precisely. The memory effect is caused by the backoff freezing regulation of IEEE 802.11 Distributed coordination function （DCF）, which has been ignored before and thus resulted in the inaccurate evaluation of the network performance. Based on our new event model, the network performance of IEEE 802.11 DCF, including throughput, packet delay distribution and energy efficiency is analyzed. Simulation results show that our model is highly accurate.     

基于效用最大化的IEEE 802.11 DCF性能分析及改进

针对802.11 DCF在系统负载较大时不能有效利用带宽资源的缺点,该文提出一种基于效用函数的DCF优化机制(U-DCF)。通过设置站点吞吐量的对数效用函数,将带宽资源的有效利用问题建模为系统效用最大化问题;应用最优化理论将此系统问题等效为可分布式求解的用户问题,即各站点只须独立选择最大化其净效用的竞争参数(CWmin),则系统整体效用也获得最大化。仿真结果表明：与标准DCF相比,U-DCF通过预估系统的当前平均分组长度和竞争站点数来调整竞争参数CWmin,能够显著提高系统的饱和吞吐量,减小分组发送时延和丢帧率。     

IEEE802.11无线局域网中一种支持业务区分的回退算法

支持QoS的MAC机制是WLAN支持QoS的关键所在.我们已经给出了一种低分组碰撞概率的MAC层回退机制——RWBO+BEB.该文进一步讨论如何让RWBO+BEB支持业务区分的问题.首先提出了一个Markov链模型,分析如何根据无线终端的带宽比率设置最小竞争窗口,然后给出了一种新的支持业务区分的回退算法——DS-RWBO,仿真结果表明,DS-RWBO能根据局域网中每个终端的带宽比率分配无线信道的带宽资源.     

一种最小竞争窗口自适应调整的802.11退避算法

该文在对原有的IEEE 802.11 DCF研究的基础上,提出了一种基于最小竞争窗口自适应调整的退避算法(Minimum Contention Window Self-adaptive Adjusting, MCWSA)。该算法的思想是每个站点根据网络中当前的时隙利用率和理论上最优时隙利用率进行比较,周期性动态调整自身的最小竞争窗口,以适应不同的网络拥塞状况。仿真表明,该算法提高了IEEE 802.11局域网在拥塞情况下的性能,在饱和吞吐量和时延上都有不同程度的改善。     

IEEE 802.11b DCF三维Markov非饱和链路模型

为了准确评估实际网络中IEEE 802.11b分布式协调功能(Distributed Coordination Function,DCF)的系统性能,提出了一种三维Markov链路模型,该模型综合考虑了DCF协议退避计数器冻结状态和有限重传次数,并结合M/M/1/K排队模型给出了在负载有限的情况下DCF系统性能的理论模型。仿真结果表明,该模型能准确地预测IEEE 802.11b DCF协议的系统性能。     

IEEE802．11DCF的一种改进退避算法

IEEE802．11中的分布式协调功能（DCF）通常采用二进制指数退避（BEB）算法。为了提高该算法的性能,在BEB算法的基础上提出了一种改进的退避算法,该算法考虑前一数据包的冲突情况,指数减小竞争窗口（CW）,并尽可能减小退避过程中的分布式帧间间隔（DIFS）开销。基于OPNET网络仿真平台,对改进算法的性能进行了仿真评估。仿真结果表明,改进后的退避算法在吞吐量和时延方面,其性能优于BEB算法和指数增加指数减小（EIED）算法。     

Performance analysis of IEEE 802.11 DCF with stochastic reward nets

In this paper, we present a performance study to evaluate the mean delay and the average system throughput of IEEE 802.11‐based wireless local area networks (WLANs). We consider the distributed co‐ordination function (DCF) mode of medium access control (MAC). Stochastic reward nets (SRNs) are used as a modelling formalism as it readily captures the synchronization between events in the DCF mode of access. We present a SRN‐based analytical model to evaluate the mean delay and the average system throughput of the IEEE 802.11 DCF by considering an on–off traffic model and taking into account the freezing of the back‐off counter due to channel capture by other stations. We also compute the mean delay suffered by a packet in the system using the SRN formulation and by modelling each station as an M/G/1 queue. We validate our analytical model by comparison with simulations. Copyright © 2006 John Wiley & Sons, Ltd.     

SCWGF:802.11 DCF竞争窗口增长因子自适应调整算法

 802.11 DCF(Distributed Coordination Function)性能的分析和改进是MAC(Medium Access Control)中一个重要的研究领域.本文在该协议的基础上提出了一种竞争窗口增长因子自适应调整算法—SCWGF(Self-adaptive algorithm of adjusting Contention Window’s Growth Factor),通过理论分析得到不同竞争节点数下竞争窗口增长因子的最优解和最优冲突时间占用率,提出了使用自适应算法使竞争窗口增长因子收敛到最优解,仿真结果表明SCWGF提高了网络的饱和吞吐量.